Panorangan, Albert Bincar (2025) Sentiment Analysis of User Reviews on the BMKG Info Application Using the Bidirectional Encoder Representations from Transformers (BERT) Method. Skripsi thesis, Universitas Pakuan.
Full text not available from this repository.Abstract
Sentiment Analysis of User Reviews on the BMKG Info Application Using the Bidirectional Encoder Representations from Transformers (BERT) Method Albert Bincar Panorangan1 , Arie Qur’ania 2 , Boldson Herdianto Situmorang3 * 1,2,3 Department of Computer Science, Faculty of Mathematics and Natural Science, Pakuan University, Bogor, West Java, 16143, Indonesia Abstract The Info BMKG app is the official platform in Indonesia that provides weather, climate, and earthquake information. Although it has been downloaded more than 5 million times, an in-depth evaluation of user reviews is essential to improve service quality. This study aims to conduct a sentiment analysis of user reviews of the Info BMKG app on the Google Play Store to classify opinions into positive, neutral, and negative categories. The method used is the pre-trained BERT-base language model with domain-adaptive pretraining through the addition of a BMKG-specific corpus. The dataset consists of 1,108 reviews collected via web scraping over two time periods between May 2023 and February 2025. The results show that the BERT-base model achieved the highest accuracy of 89% with the following hyperparameter configuration: batch size 32, learning rate 2e-5, and 15 epochs. The addition of the BMKG domain corpus was proven to enhance the model’s ability to understand specific contexts related to meteorological and geophysical information. This study concludes that the BERT-base method is highly effective for sentiment analysis of public service app reviews in Indonesia. Keywords: Sentiment Analysis, BMKG Information, BERT, Natural Language Processing, Google Play Store
| Item Type: | Thesis (Skripsi) |
|---|---|
| Subjects: | Fakultas Ilmu Pengetahuan Alam dan Matematika > Ilmu Komputer |
| Divisions: | Fakultas Matematika dan Ilmu Pengetahuan Alam > Ilmu Komputer |
| Depositing User: | PERPUSTAKAAN FAKULTAS MATEMATIKA DAN ILMU PENGETAHUAN ALAM UNPAK |
| Date Deposited: | 08 Jul 2026 07:45 |
| Last Modified: | 08 Jul 2026 07:45 |
| URI: | http://eprints.unpak.ac.id/id/eprint/10809 |
Actions (login required)
 |
View Item |